Binary Audio Classification on CREMA-D (test)
72.45Mean AccuracyAdaptive Attribute Discovery with MLLMs
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Adaptive Attribute Discovery with MLLMsMethod Type=Proposed, Backbone=gemini-3-f-p, Trials=10, Sampling Temperature=1.0, q=10, T=10, k=10, Weak Classifier=decision stump2026.03 | 72.45 | 6.1 | 3.45 | |
| LRMethod Type=Baseline, Features=CLAP [18], Solver=L-BFGS2026.03 | 70 | — | — | |
| MLLM PredictionMethod Type=Baseline, Backbone=gemini-3-f-p, Trials=10, Sampling Temperature=1.0, Evaluation Protocol=direct zero-shot prediction2026.03 | 69 | 2.32 | — |